Monitoring Of Arid Land Soil Salinization Using Remote Sensing And Construction Of Early Warning Network Transmission System

Soil salinization is one of the land degradation types in arid and semi arid regions. The global extent of primary salt-affected soils is about955M ha, while secondary salinization affects some77M ha, with58%of these in irrigated areas. The damage of soil salinization to the economy, environment and agricultural ecological system is concerned by some scholars year by year. This requires careful monitoring of the soil salinity status and variation to curb degradation trends, and secure sustainable land use and management. It is very important to study the quality, the scope, the area, the distribution, the dynamic change and the degree of saline soils. The combination of remote sensing with some traditional methods has some important meaning for studying the characteristics, areas, degree, distribution, time and space variation, management and protecting from enlarging of soil salinization. This study aiming at the problems mentioned above, firstly, using time sequence remote sensing image and field measured data, analyze the quantify relationship between the salinization degree and land surface parameters, propose a new index-salinization detecting index (SDI) which can comprehensively reflect salinized soil biophysical features; point out salinized soil information extracting method which based on remote sensing image fusion; establish object-oriented soil salinization information extracting method. Secondly, analyze the problems of soil salinization time and space evolution, early warning and alarming,, evaluate the soil salinization situation at the study area. At the end, using computer network and Web GIS techniques, develop monitoring of the soil salinization using remote sensing and early warning network transmission system.(1) The object-oriented classification method was used to extract soil salinization information from the ALOS images, and the results were compared with the traditional information extraction method. The results show that the overall accuracy was89.38%, with a kappa coefficient value of0.88of this method. And the overall accuracy and kappa coefficient value were increased8.24%and0.08, respectively, compared with the traditional method; the object-oriented method is superior to traditional method in high resolution remote sensing information extraction.(2) The ALOS Panchromatic image and Multi-spectral images are selected for the image fusion method, including HIS, PCA, Gram-Schmidt and Wavelet. The performance of each method was analysed by qualitatively and quantitatively.Also supervised classification method used to extract the soil salinization Information from multispectral images and fused images respectively. The results show that the Wavelet fusion method can provide more information, higher spatial resolution and higher classification resultsis accuracy compared with multi-spectral images.It is most effective image fusion method for monitoring the salinized soil, an advantage over the common data fution methods such as HIS, PCA and Gram-Schmidt.(3) The relationship between the salinization degree and land surface parameters such as NDVI and albedo, was analyzed quantitatively, using Landsat ETM+and field measured data.The salinization detecion index (SDI) is developed.(4) Using4temporal images combined with field survey data. With the application of Remote Sensing and Geographic Information Systems, classified statistic and LUCC transform matrix were acquired. In the period from1989to2001, the areas of farmland,woodland, moderate saline land, water and the other (Gobi, sand and clay, etc.) have increased; slight saline land and heavy saline land has reduced. In the period from2001to2006, the areas of slight saline land, moderate saline land and heavy saline land have increased; the other types of land use/covers have reduced. In the period from2006to2010, the areas of moderate saline land, heavy saline land, water and the other have increased; the other types of land use/covers have reduced; As time went by, the gravity center of every type of land use/covers has changed. The farthest distance is the change of gravity center of water between1989and2001, which is23.12km. At the same period, the distance of the gravity center of heavy saline land moved to the northwest is smaller than that of water, which is22.04km. The gravity center of woodland moved to the southwest, and the transferred distances increased year by year. The gravity center of moderately and slightly saline land in essence transferred to northwest during these21years.(5) Under several important principles guidance, using Web GIS technique, the salinization early warning system was made in research area. The system development by the plat-form of Arc GIS server, SQL DBMS, combined with Net. This is very important for decision making for the relevant department.